False Positives on Motion Detection

[FarmerCharlie 0]

I have been noticing a lot of false alarms that appear to be caused by glitches on some of the cameras. The glitches look a little like brief loss of sync or some other glitch that appears as motion to the motion detection software in the DVR card. The card is an Avermedia NV3000.

 

I have two cameras that use the inexpensive 12 volt power supplies that came with the cameras plus one Sony TRV900 camera (Camera 1) that has it's own power supply plus battery. The alarms seem to be mostly on the cameras on the inexpensive power supplies.

 

The three attached images illustrate one example. Frame 2 shows a vertical jump on camera 2.

 

Are these types of glitches typical? Would I likely see improvement if I used a better multichannel power supply for the cameras. Or would a UPS likely help? I am not getting power glitches sufficient to cause the computer to recycle.

 

Thanks,

Charles

[mateck8888 0]

First of all, looking at your video subject, can I assume you are working for Homeland Security? Or is this project being done with stimulus funds?!?!

 

What kind of coax are you using to connect the cheaper cameras to your Aver? If you have the really cheap thin wire stuff that may be part of your problem, especially if you have them running a distance with splice connectors in between. Also, you may be right on the power supply. Try taking a volt meter, and read the power at the camera side to see what you are actually getting. If it is an unregulated power supply, you may have to do this while the camera is drawing current.

 

Also, it is possible that some of your problem is simply noise in the picture. It may be recording when it sees low light noise, in addition to your vertical bar roll.

[FarmerCharlie 0]

mateck8888,

 

Thanks for the suggestions.

To satisfy your fears, I am not employed by Homeland Security; but if some bad guys happen to come messing around with my purple martins in the dark of night, you can bet that I will sic the US Fish and Wildlife Service on them for messing around with protected migratory birds. And the only connection with the Stimulus is my substantial financial support of the whole spy and CCTV industry this year with my bird obsession.

 

The cables I am using are the fairly inexpensive cables that either came with the cameras or that I ordered from SpyTown and other sources. They are not the sort of quality that I used to use in video production, but they are marketed for this type of application. Also, I am using the same cables with the Sony TRV900, which appears to have fewer alarms..

 

I agree that some of the false alarms are caused by ordinary noise, leaves blowing in the wind, or moths that flew up close to the IR lights that were originally mounted on the distant camera. But the ones like the images I posted with an obvious glitch seem to be the cause of many. I don't expect to eliminate all false positives with cameras of this quality with varied lighting and weather, and moths. I just want to make it practical to playback the videos at high speed to check for predators (of the 4-legged and winged type).

 

I will need to make a Y adapter to measure the voltage under load. That sounds like a good thing to do. Do you think getting one of the fairly inexpensive multichannel regulated power supplies would be a reasonable investment?

 

Thanks,

Charles

[mateck8888 0]

Generally speaking, most of the short run cables that come with these cameras are really very thin, with little shielding compared to RG-59 or RG-6 coax. For example, if you buy a DVD player it will come with cheap patch cords that would be fine for video editing, as they are only 3 feet long. When you get into long runs, you should be using a thick black cable that says RG-6 or RG-59 on it.

 

But I'm not entirely sure that is your problem. The voltage reading might be more interesting. Also, it is always good to test these systems on the bench. Hook up your two offending cameras with a 6 foot patch cord, and try recording for 24 hours with full room light with nothing room light. Then see how many "glitches" you get.

 

Anyway, before you get a regulated power supply, it would be interesting to see what you are pulling now, and how things work out in a bench test, if you go that far with the troubleshooting.

 

What a great use for a TRV900. I paid $2000 for mine back in 1999, and that thing was a real workhorse. I think it was the smallest 3-chip camera of its time.

 

Matt

[FarmerCharlie 0]

What a great use for a TRV900. I paid $2000 for mine back in 1999, and that thing was a real workhorse. I think it was the smallest 3-chip camera of its time.

 

Matt

I debated long and hard back in 1999 or so between the TRV900 and the CanonXL1. Finally went on portability rather than interchangeable lenses. That was probably the best decision then, but I sure wish I could put a Canon telephoto on it and eliminate the Sony digital zoom.

 

I'll test your suggestions. I have plenty of time between now and the time the martins return next year. I'm still streaming to the Internet at http://www.purplemartin.tv/charles_branch_combined/ but it's pretty dull now that my birds have left for winter vacation in Rio.